This blog is co-authored by Wee Hyong Tok, Principal Data Scientist Manager, Office of the CTO AI.
In recent years, we have seen a leap in practical AI innovations catalyzed by vast amounts of data, the cloud, innovations in algorithms, hardware and more. So how do developers begin to design AI applications that engage and delight your customers, optimize operations, empower your employees, and transform products?
Using Azure Cognitive Services you can now infuse your applications, websites, and bots with intelligent capabilities. These capabilities build on years of research done on vision, speech, knowledge, search, and language. Using different cognitive services, developers can now easily add AI capabilities without training the machine learning models from scratch.
O’Reilly and Microsoft are excited to bring you a free e-book on AI, titled A Developer’s Guide to Building AI Applications. In this e-book, Anand Raman and Wee Hyong Tok of Microsoft provide a gentle introduction to use Azure AI for building intelligent, AI applications. They provide a practical example of a bot called “Conference Buddy”, that is used by conference attendees. The e-book walks through the use case, the architecture, and how to create the bot while infusing it with AI. The code
Today’s consumers are using more devices and channels to interact with retailers than ever before. Seamless service across all channels is the expectation, not the exception. Digital-first brands are raising the bar for hyper-personalized experiences through their test-and-learn approach to rapid delivery of modern commerce capabilities. As we all know, traditional brick and mortar retailers are under significant pressure due to high fixed costs associated with managing traditional store infrastructure, material decreases, and flat year-over-year comp-store revenue. A consequence of this pressure is the inability to innovate and create new, competitive user experiences. One answer to this problem is the introduction of a new service on the Azure Marketplace.
You say monoliths, I say microservices.
It’s estimated that more than half of retailers still operate their businesses on monolithic, on-premises commerce applications. Those monoliths inhibit speed and flexibility to build hyper-personalized experiences. They lack agility to support new business models to differentiate. Other drawbacks include all night deployments, six months from concept to go-live, 24 hours to deploy one line of code change, and the requirement of significant rollout and readiness planning.
Unless retailers revamp their platforms, they will be challenged to keep up with new competitors who are operating
Healthcare is drowning in data. Every patient brings a record that could span decades, with x-rays, MRIs, and other data that can affect every decision. Providers and payers bring their own collateral to the table. Skills, policies, and certifications are just the start. Complying with regulations such as the Health Insurance Portability and Accountability Act (HIPAA) present additional burdens on any process. But there is a remedy: AI (Artificial Intelligence) and ML (machine learning) are powerful tools that can change the way healthcare organizations deal with the tsunami of healthcare data.
Use cases where AI can help vary from diagnostic imaging to predicting the patient length of stay, and even chatbots. But these initiatives must avoid breaches, ransomware, and other privacy compliance issues.
Join this upcoming webinar live on October 18, 2018, at 11:00 AM Pacific Time, bringing you thought leaders from across the healthcare industry including:
Andrew Hicks, VP, Healthcare Assurance Services at Coalfire Mitch Parker, Executive Director, Information Security, and Compliance at IU Health David Houlding, Principal Healthcare lead at Microsoft
Each presenter will provide practical advice on how to address security, privacy, and compliance with AI solutions in healthcare.
Key takeaways: AI in healthcare trends and risks
Combining biometric identification with artificial intelligence (AI) enables banks to take a new approach to verifying the digital identity of their prospects and customers. Biometrics is the process by which a person’s unique physical and personal traits are detected and recorded by an electronic device or system as a means of confirm identity. Biometric identifiers are unique to individuals, so they are more reliable in confirming identity than token and knowledge-based methods, such as identity cards and passwords. Biometric identifiers are often categorized as physiological identifiers that are related to a person’s physicality and include fingerprint recognition, hand geometry, odor/scent, iris scans, DNA, palmprint, and facial recognition.
But how do you ensure the effectiveness of identifying a customer when they are not physically in the presence of the bank employee? As the world of banking continues to go digital, our identity is becoming the key to accessing these services. Regulators require banks to verify that users are who they say they are, not bad actors like fraudsters or known money launderers. And verifying identities online without seeing the person face to face is one of the biggest challenges online and mobile services face today.
It’s problematic because identity documents
Many healthcare organizations are starting to adopt artificial intelligence (AI) systems to gain deeper insight into operations, patient care, diagnostic imaging, cost savings and so on. However, it can sometimes be daunting to even know where to get started. Many times, you need a clear lighted path to start your journey and embrace AI and machine learning (ML) capabilities rapidly.
One method is using an Azure Healthcare AI blueprint. It’s a shortcut to using Microsoft Azure at low cost and without deep knowledge of cloud computing. Blueprints include resources such as example code, test data, security, and compliance support. The largest advantage of using a blueprint is explicit advice and clear instructions on keeping your solution in compliance. We’re trying to eliminate the mystery, so you don’t have to research it yourself.
Three core areas where the blueprint can help with compliance are cloud provider and client responsibilities, security threats, and regulatory compliance. These three areas can get overlooked at the beginning of any technology project, yet they are important parts of creating healthcare systems. Applying formal discipline to these areas is made easier by using the blueprint to create an AI/ML experiment installation.
The blueprint includes
Devices and technologies are moving forward at a rapid pace, though the everyday tools we use remain relatively unchanged. What if we could infuse AI into everyday tools to delight and inspire developers to do more using Microsoft AI platform? With just a little bit of creativity and using Microsoft’s current AI offerings, we can bring AI capabilities closer to customers and create applications that will inspire every organization, every developer, and every person on this planet.
Introducing Snip Insights
An open source cross-platform AI tool for intelligent screen capture. Snip Insights revolutionizes the way users can generate insights from screen captures. The initial prototype of Snip Insights, built for Windows OS and released at Microsoft Build 2018 in May, was created by Microsoft Garage interns based out of Vancouver, BC. Our team at Microsoft AI Lab in collaboration with the Microsoft AI CTO team took Snip Insights to the next level by giving the tool a new intuitive UX, cross-platform availability (MacOS, Linux, and Windows), and free download and usage under MSA license. Snip Insights leverages Microsoft Azure’s Cognitive Services APIs to increase users’ productivity by reducing the number of steps needed to gain intelligent insights.
Frost and Sullivan has estimated the Artificial Intelligence (AI) Market for Healthcare IT in 2018 for hospitals at $574 million USD and expects it to grow at a CAGR of 65 percent to $4.3 billion USD in 2022! Similar explosive growth is predicted across other segments of healthcare. Machine Learning (ML) is a type of AI that has already seen successful application in healthcare, particularly rapid growth, and shows major untapped potential to further help improve healthcare going forward. Key use cases range from resource and asset optimization to readmission prevention, chatbots, anti-fraud, behavioral analytics, medical risk analytics, claims analytics, cybersecurity, and many more. Business values driving healthcare organizations to deploy AI solutions across these use cases span cost reduction, improving patient outcomes, and improving the engagement and experiences of patients and healthcare professionals. Major opportunities for startups range from creating AI products and solutions for specific use cases and healthcare needs to services for education, customizing AI solutions, integrating them with existing enterprise systems and data stores, managing and operating solutions, and so forth. See the AI in Healthcare Guide for more information on these use cases and opportunities.
Below I review key goals of any healthcare AI startup,
Organizations around the world are gearing up for a future powered by data, cloud, and Artificial Intelligence (AI). This week at Spark + AI Summit Europe, I talked about how Microsoft is committed to delivering cutting-edge innovations that help our customers navigate these technological and business shifts.
The driving force behind powerful AI applications is data – and getting the most out of AI requires a modern data estate. Organizations are using their data to extract important insights to drive their businesses forward and engage their customers in ways that were simply not possible before. One such example is the Real Madrid Football Club, one of the world’s top sports franchises with 500 million fans worldwide. Real Madrid built a global digital sports platform to engage one-on-one with fans, implement personalized promotional campaigns, and use data to track and analyze fan behaviors, among many other capabilities. This data-driven strategy has led to a 400 percent increase in installed fan base, and a 30 percent increase in digital revenue growth for the club.
“We used to pull data from just five sources before, but now we pull from more than 70 sources using our Microsoft Azure platform. This has enabled
Today we are excited to strengthen our commitment to supporting PyTorch as a first-class framework on Azure, with exciting new capabilities in our Azure Machine Learning public preview refresh. In addition, our PyTorch support extends deeply across many of our AI Platform services and tooling, which we will highlight below.
During the past two years since PyTorch’s first release in October 2016, we’ve witnessed the rapid and organic adoption of the deep learning framework among academia, industry, and the AI community at large. While PyTorch’s Python-first integration and imperative style have long made the framework a hit among researchers, the latest PyTorch 1.0 release brings the production-level readiness and scalability needed to make it a true end-to-end deep learning platform, from prototyping to production.
Four ways to use PyTorch on Azure Azure Machine Learning service
Azure Machine Learning (Azure ML) service is a cloud-based service that enables data scientists to carry out end-to-end machine learning workflows, from data preparation and training to model management and deployment. Using the service’s rich Python SDK, you can train, hyperparameter tune, and deploy your PyTorch models with ease from any Python development environment, such as Jupyter notebooks or code editors. With Azure ML’s deep
At a previous position, I owned the software and hardware testing across a 6,000-branch network for a large fortune 100 bank in the U.S. The complexity and sophistication of the end-to-end delivery of products and services to existing customers was daunting. Vetting new, potential customers while simultaneously building a new system was tough. Especially since the new system had to balance a pleasant front-end user experience with the backend processes from strong Know Your Customer (KYC) scrubbing (in other words, due diligence). That backend system was invisible, batch-based, and only had post-transaction look back capability. I learned that banks can have their cake and eat it too, but the business implications of limiting user friction are not trivial, and properly vetting customers puts a lot of pressure on the technology capabilities.
As a compliment to my online and mobile fraud theme this quarter see Detecting Online and Mobile Fraud with AI, I’ll provide some insights on how banks are seeking to rationalize and simplify security and compliance processes in real-time. The path stretches from the device to the network and back-end infrastructure. The goal is to ease the burden on employees and reduce costs from fines. Specifically, the requirements